Purpose This study investigated the associations between athletic and exercise identities and a range of disordered eating and exercise behaviors. Method Study 1 (N = 441; 53.5% female) included online questionnaires of athletic identity and disordered eating behaviors, anthropometrics, and demographic information. Study 2 (N = 225; 50.2% female) added measures of: exercise identity, exercise behaviors, and a broader measure of disordered eating behaviors. Multiple linear regressions were used to test hypotheses. Dominance analysis was used in Study 2 to determine the unique variance accounted for by each predictor. Results Controlling for body mass index (BMI), age, and sex, athletic identity was not significantly associated with disordered eating in Study 1 (ps > 0.039) but was significantly associated with lower cognitive restraint in Study 2 (p = 0.012). In Study 2, stronger exercise identity was significantly associated with more cognitive restraint (p < 0.001), more body dissatisfaction (p = 0.016), more compulsive exercise (p < 0.001), and more positive and healthy exercise (p < 0.001), after controlling for BMI, age, sex, and athletic identity. Dominance analyses suggested that exercise identity was more strongly associated with these outcomes than athletic identity. Conclusion Exercise identity, but not athletic identity, may be an important risk factor for disordered eating and exercise behaviors in broad populations.